Schedule¶
| DAY 1 - Tuesday 26/11 | |
|
09:00 CET
10:00 EET |
Welcome and Introduction
Presenters: Jørn Dietze (LUST) |
|
09:15 CET
10:15 EET |
Introduction to LUMI
Presenter: Jørn Dietze (LUST) |
|
09:45 CET
10:45 EET |
Using the LUMI web interface
Presenters: Mats Sjöberg (CSC) and Lukas Prediger (CSC) |
|
10:05 CET
11:05 EET |
Hands-on: Run a simple PyTorch example notebook |
|
10:35 CET
11:35 EET |
Break (15 minutes) |
|
10:50 CET
11:50 EET |
Your first AI training job on LUMI
Presenters: Mats Sjöberg (CSC) and Lukas Prediger (CSC) |
|
11:20 CET
12:20 EET |
Hands-on: Run a simple single-GPU PyTorch AI training job |
|
12:05 CET
13:05 EET |
Lunch break (45 minutes) |
|
12:50 CET
13:50 EET |
Understanding GPU activity & checking jobs
Presenter: Samuel Antão (AMD) |
|
13:10 CET
14:10 EET |
Hands-on: Checking GPU usage interactively using rocm-smi |
|
13:30 CET
14:30 EET |
Running containers on LUMI Presenter: Jan Vicherek (LUST) |
|
13:50 CET
14:50 EET |
Hands-on: Pull and run a container |
|
14:10 CET
15:10 EET |
Break (15 minutes) |
|
14:25 CET
15:25 EET |
Building containers from conda/pip environments Presenter: Jørn Dietze (LUST) |
|
14:45 CET
15:45 EET |
Hands-on: Creating a conda environment file and building a container using cotainr |
|
15:05 CET
16:05 EET |
Extending containers with virtual environments for faster testing Presenter: Gregor Decristoforo (LUST) |
|
15:25 CET
16:25 EET |
Hands-on: Getting started with your own project
Bring your own AI code, you want to run on LUMI, and spent some time applying what you have learned during the workshop - with on-site support from LUST/AMD. |
|
16:10 CET
17:10 EET |
End of the course day |
| DAY 2 - Wednesday 27/11 | |
|
09:00 CET
10:00 EET |
Scaling AI training to multiple GPUs
Presenters: Mats Sjöberg (CSC) and Lukas Prediger (CSC) |
|
09:40 CET
10:40 EET |
Hands-on: Converting the PyTorch single GPU AI training job to use all GPUs in a single node via DDP |
|
10:20 CET
11:20 EET |
Break (15 minutes) |
|
10:35 CET
11:35 EET |
Extreme scale AI
Presenter: Samuel Antão (AMD) |
|
11:15 CET
12:15 EET |
Demo/Hands-on: Using multiple nodes |
|
11:30 CET
12:30 EET |
Loading training data on LUMI
Presenter: Harvey Richardson (AMD) |
|
12:00 CET
13:00 EET |
Lunch break (60 minutes) |
|
13:00 CET
14:00 EET |
Coupling machine learning with HPC simulation Presenter: Harvey Richardson (HPE) |
|
13:30 CET
14:30 EET |
Hands-on: Advancing your own project and Q&A
Bring your own AI code, you want to run on LUMI, and spent some time applying what you have learned during the workshop - with on-site support from LUST/AMD. |
|
15:45 CET
16:45 EET |
End of the course day |